Non-alcoholic steatohepatitis (NASH) is a chronic liver disease that occurs in individuals without significant alcohol consumption and histologically it resembles alcoholic liver disease with macrovesicularsteatosis, spotty necrosis, inflammation, Mallory bodies, and fibrosis. It is increasingly being recognized as a predominant type of chronic liver disease in the United States; however, its prevalence, pathogenesis, and natural history have not been adequately studied. In order to better understand the epidemiology and pathogenesis and to identify optimal therapy for NASH, we propose to conduct NASH-related clinical research in the following specific aims: Specific Aim 1: The objective is to create a database of patients with fatty liver and NASH that will enable us to perform multidisciplinary and multicentered studies on the epidemiology, pathogenesis, and therapy of NASH. The subgroups of patients included in this database are adults with fatty liver and NASH, children (< 18yrs of age) with fatty liver and NASH, women with polycystic ovarysyndrome, and appropriately matched controls. The cohort will be characterized clinically, anthropometrically, through laboratory tests, and histologically. A repository containing liver tissue, blood samples, and DNA from subjects with disease and controls will be developed; Specific Aim 2:The overall goal of this specific aim is to derive and validate risk equations that measure and stratify the risk for advanced histology in patients with non- alcoholic fatty liver disease. To this end, we will conduct a multicenter, nested case-control study of patients with simple fatty-liver, non-cirrhotic NASH, and cirrhotic NASH to identify risk factors for advanced histology, and to derive and validate a clinical prediction rule for reliable identification of those with advancedhistology; Specific Aim 3: We hypothesize that insulin resistance is pivotal in the pathogenesis of NASH and measures that improve insulin resistance would lead to an improvement in liver histology in non-diabetic NASH. Here, we propose to conduct a multicenter, randomized, double-blind, placebo-controlled study of moderate weight reduction with or without metformin for 12-months. The primary end-point is the change in liver histology measured by comparing biopsy findings at baseline and at the end of 12-month treatment period. The secondary end-points are changes in insulin resistance, lipid peroxidation, determinants of oxidative stress, anthropometric measurements, liver biochemistry, and the measures of quality of life.